9:05 a.m.
Hi,
Sometimes once per a few times, I faced the following weird build
error when building Ruby modules.
Did you see it? Known issue?
https://koji.fedoraproject.org/koji/taskinfo?taskID=48194471
BuildrootError: could not init mock buildroot, mock exited with status
20; see root.log for more information
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/root.log
=> looks ok.
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/build.log
=> empty.
Thanks.
--
Jun | He - His - Him
5:21 p.m.
On Fri, Jul 31, 2020 at 04:05:20PM +0200, Jun Aruga wrote:
Hi,
Sometimes once per a few times, I faced the following weird build
error when building Ruby modules.
Did you see it? Known issue?
https://koji.fedoraproject.org/koji/taskinfo?taskID=48194471
BuildrootError: could not init mock buildroot, mock exited with status
20; see root.log for more information
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/root.log
=> looks ok.
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/build.log
=> empty.
Weird. The actual error is:
"ERROR: Could not find useradd in chroot, maybe the install failed?"
But that sounds like what happens when it didn't install the build rpms
in there.
Can you try again now? we fixed a major issue with mbs, which I wouldn't
think was related to this, but who knows...
kevin
11:38 a.m.
Hi,
Weird. The actual error is:
"ERROR: Could not find useradd in chroot, maybe the install failed?"
But that sounds like what happens when it didn't install the build rpms
in there.
Can you try again now? we fixed a major issue with mbs, which I wouldn't
think was related to this, but who knows...
Thanks for checking. Ah I found the error message you mentioned in
mock_output.log with "Option --old-chroot has been deprecated.".
Weird. I just executed the command `fedpkg module-build` without
`--old-chroot` or any options.
```
$ rpm -q fedpkg
fedpkg-1.38-4.fc32.noarch
```
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/mock_output.log
```
ERROR: Option --old-chroot has been deprecated. Use --isolation=simple instead.
...
ERROR: Could not find useradd in chroot, maybe the install failed?
```
Sure. Let me try it.
--
Jun | He - His - Him
5 a.m.
Dne 02. 08. 20 v 0:21 Kevin Fenzi napsal(a):
On Fri, Jul 31, 2020 at 04:05:20PM +0200, Jun Aruga wrote:
> Hi,
> Sometimes once per a few times, I faced the following weird build
> error when building Ruby modules.
> Did you see it? Known issue?
>
> https://koji.fedoraproject.org/koji/taskinfo?taskID=48194471
> BuildrootError: could not init mock buildroot, mock exited with status
> 20; see root.log for more information
>
> https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/root.log
> => looks ok.
> https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/build.log
> => empty.
>
Weird. The actual error is:
"ERROR: Could not find useradd in chroot, maybe the install failed?"
It seems there is nothing installed into build root except of
module-build-macros-0.1-1.module_f30+8976+2502a8ed.noarch, that does not
appear to be enough to have `useradd` present in the build root.
So isn't this the relevant part of root.log which could give a hint?
~~~
DEBUG util.py:621: No match for group package "rpm-build"
DEBUG util.py:621: No match for group package "fedpkg-minimal"
DEBUG util.py:621: No match for group package "shadow-utils"
DEBUG util.py:621: No match for group package "gnupg2"
DEBUG util.py:621: No match for group package "bash"
DEBUG util.py:621: No match for group package "glibc-minimal-langpack"
DEBUG util.py:621: No match for group package "redhat-rpm-config"
DEBUG util.py:621: No match for group package "fedora-release"
~~~
Vít
But that sounds like what happens when it didn't install the build rpms
in there.
Can you try again now? we fixed a major issue with mbs, which I wouldn't
think was related to this, but who knows...
kevin
_______________________________________________
devel mailing list -- devel(a)lists.fedoraproject.org
To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org
9:47 a.m.
Can you try again now? we fixed a major issue with mbs, which I
wouldn't
think was related to this, but who knows..
I tried it, and found the cause of the error now.
So isn't this the relevant part of root.log which could give a
hint?
Yes, possibly it is related to the issue.
https://koji.fedoraproject.org/koji/taskinfo?taskID=48194471
BuildrootError: could not init mock buildroot, mock exited with status
20; see root.log for more information
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/root.log
=> looks ok.
https://kojipkgs.fedoraproject.org//work/tasks/4473/48194473/build.log
=> empty.
The situation is when I got the above error, I got the 4 builds by
`fedpkg module-build`.
(I am excluding el8 by `platform: [-el8]`) in the module config file ruby.yaml.
```
$ fedpkg module-build
Submitting the module build...
The builds 9293, 9294, 9295 and 9296 were submitted to the MBS
```
The error happened on build id 9293.
But today I got the 3 builds with the same config file except the build id 9293.
```
$ fedpkg module-build
Submitting the module build...
The builds 9296, 9294 and 9295 were submitted to the MBS
```
I found `ruby-None-None` in the build id 9293.
So, I suppose previously it tried to build outputed platform (f30).
The other build id 9294 is f32, 9295 is f31, 9296 is f33.
```
$ fedpkg module-build-info 9293
...
Name: ruby
NVR: ruby-None-None
State: FAILED
Koji Task: https://koji.fedoraproject.org/koji/taskinfo?taskID=48194471
...
```
--
Jun | He - His - Him
10:49 a.m.
On Mon, Aug 03, 2020 at 04:47:39PM +0200, Jun Aruga wrote:
...
So, I suppose previously it tried to build outputed platform (f30).
The other build id 9294 is f32, 9295 is f31, 9296 is f33.
Ah, that would do it. I finally was shown the command to retire the f30
eol platform and did so... before it was likely trying to build against
it, but those targets are all gone, so nothing was in the build but the
mbs macro package. ;(
kevin
1369
days inactive
1372
days old
5 comments
3 participants
participants (3)
-
Jun Aruga -
Kevin Fenzi -
Vít Ondruch